Copertina

Connecting the Dots: LLMs can Infer and Verbalize Latent Structure from Disparate Training Data

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Treutlein, Johannes, Choi, Dami, Betley, Jan, Marks, Samuel, Anil, Cem, Grosse, Roger, Evans, Owain
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Computation and Language Artificial Intelligence Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

School of Reward Hacks: Hacking harmless tasks generalizes to misaligned behavior in LLMs
di: Taylor, Mia, et al.
Pubblicazione: (2025)

The Consciousness Cluster: Emergent preferences of Models that Claim to be Conscious
di: Chua, James, et al.
Pubblicazione: (2026)

Conditional misalignment: common interventions can hide emergent misalignment behind contextual triggers
di: Dubiński, Jan, et al.
Pubblicazione: (2026)

Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs
di: Betley, Jan, et al.
Pubblicazione: (2025)

Tell me about yourself: LLMs are aware of their learned behaviors
di: Betley, Jan, et al.
Pubblicazione: (2025)

Subliminal Learning: Language models transmit behavioral traits via hidden signals in data
di: Cloud, Alex, et al.
Pubblicazione: (2025)

Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
di: Betley, Jan, et al.
Pubblicazione: (2025)

Thought Crime: Backdoors and Emergent Misalignment in Reasoning Models
di: Chua, James, et al.
Pubblicazione: (2025)

Me, Myself, and AI: The Situational Awareness Dataset (SAD) for LLMs
di: Laine, Rudolf, et al.
Pubblicazione: (2024)

Lessons from Studying Two-Hop Latent Reasoning
di: Balesni, Mikita, et al.
Pubblicazione: (2024)

Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers
di: Karvonen, Adam, et al.
Pubblicazione: (2025)

Are DeepSeek R1 And Other Reasoning Models More Faithful?
di: Chua, James, et al.
Pubblicazione: (2025)

Influential Training Data Retrieval for Explaining Verbalized Confidence of LLMs
di: Xia, Yuxi, et al.
Pubblicazione: (2026)

Connecting the Dots: Inferring Patent Phrase Similarity with Retrieved Phrase Graphs
di: Peng, Zhuoyi, et al.
Pubblicazione: (2024)

Why and How LLMs Hallucinate: Connecting the Dots with Subsequence Associations
di: Sun, Yiyou, et al.
Pubblicazione: (2025)

Predictive Concept Decoders: Training Scalable End-to-End Interpretability Assistants
di: Huang, Vincent, et al.
Pubblicazione: (2025)

Can We Infer Confidential Properties of Training Data from LLMs?
di: Huang, Pengrun, et al.
Pubblicazione: (2025)

Can Language Models Explain Their Own Classification Behavior?
di: Sherburn, Dane, et al.
Pubblicazione: (2024)

Inferring the presence and abundance of rare waterbirds species from scarce data
di: Bricout, Barbara, et al.
Pubblicazione: (2026)

Connecting the Dots in News Analysis: Bridging the Cross-Disciplinary Disparities in Media Bias and Framing
di: Vallejo, Gisela, et al.
Pubblicazione: (2023)

Training Data Attribution via Approximate Unrolled Differentiation
di: Bae, Juhan, et al.
Pubblicazione: (2024)

Cylinder decompositions on geometric armadillo tails
di: Lee, Dami, et al.
Pubblicazione: (2024)

ESTRUTURA DE PROPRIEDADE NO BRASIL: EVIDÊNCIAS EMPÍRICAS NO GRAU DE CONCENTRAÇÃO ACIONÁRIA
di: Anamélia Borges Tannus Dami
Pubblicazione: (2007)

Estrutura de propriedade no Brasil: Evidências empíricas no grau de concentração acionária
di: Anamélia Borges Tannus Dami
Pubblicazione: (2023)

Discrete Vector Bundles with Connection
di: Berwick-Evans, Daniel, et al.
Pubblicazione: (2021)

Resonant Structures in $p{}^7\mathrm{Be}$ Scattering and Their Connection to the Astrophysical $S$-Factor
di: Khachi, Anil
Pubblicazione: (2025)

The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets
di: Marks, Samuel, et al.
Pubblicazione: (2023)

Kinetic analysis of phase transformations during continuous heating: Crystallization of glass-forming liquids
di: Houghton, Owain S.
Pubblicazione: (2025)

ALAS: Measuring Latent Speech-Text Alignment For Spoken Language Understanding In Multimodal LLMs
di: Mousavi, Pooneh, et al.
Pubblicazione: (2025)

Bayesian Tensor Decomposition for Clustering Latent Symptom Profiles for Verbal Autopsy Data
di: Yu Zhu, et al.
Pubblicazione: (2026)

On Verbalized Confidence Scores for LLMs
di: Yang, Daniel, et al.
Pubblicazione: (2024)

Effective Faraday interaction between light and nuclear spins of Helium-3 in its ground state: a semiclassical study
di: Fadel, Matteo, et al.
Pubblicazione: (2024)

Beyond Data Filtering: Knowledge Localization for Capability Removal in LLMs
di: Shilov, Igor, et al.
Pubblicazione: (2025)

The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A"
di: Berglund, Lukas, et al.
Pubblicazione: (2023)

Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
di: Hubinger, Evan, et al.
Pubblicazione: (2024)

The growth of the mussel Mytilus californianus
di: Richards, Owain Westmacott
Pubblicazione: (1928)

Put Aside Your Pencil: How Talk Becomes Writing Through Verbal Rehearsal
di: Kristen I. Evans
Pubblicazione: (2026)

Connecting the Dots: Training-Free Visual Grounding via Agentic Reasoning
di: Luo, Liqin, et al.
Pubblicazione: (2025)

Unveiling the Limits of Large Language Models in Inferring Pragmatic Meaning from Non-Verbal Responses
di: Eo, Sugyeong, et al.
Pubblicazione: (2026)

Negation Neglect: When models fail to learn negations in training
di: Mayne, Harry, et al.
Pubblicazione: (2026)